7/30/2019 Surface Degradation of Ductile Metals in Elevated http://slidepdf.com/reader/full/surface-degradation-of-ductile-metals-in-elevated 1/14 Wear, 111 (1986) 173 - 186 173 SURFACE DEGRADATION OF DUCTILE METALS IN ELEVATED TEMPERATURE GAS-PARTICLE STREAMS ALAN LEVY and YONG-FA MAN Materi als and M olecular Research Division, L awrence Berkeley Laboratory, Berkeley, CA 94720 (U.S.A.) (Received February 15, 1985; revised November 5, 1985; accepted December 20,1985) Summary The mechanisms and rates of erosion and combined erosion-corrosion of SCr-1Mo steel (where the compositi on is in approximate weight per cent) and type 310 stainless steel at elevated temperatur es were investigated to understand better the behavior of piping steels in fluidized bed combustor environments. Tests were performed in a partially inert gas atmosphere to study erosion behavior and in an air atmosphere to study combined erosion- corrosion behavior. I t was determined that the erosion rate remained con- stant or decreased wi th increasing temperature in nitrogen unti l a temper- atur e was reached at which the tensil e strength started to decrease more rapidly with increasing test temperatur e. Above this temperatur e the erosion rate increased rapidly with temperature. I n an erosion-corrosion environment, corrosion was the. dominant mechanism at all test conditi ons. At higher temperatur es and velocities the material loss mechanism changed from low loss rate chipping of the scale to high loss rate periodic spalling. The continuous scale formed on SCr-1Mo steel in air appeared to protect the metal surface, decreasing its loss rate in (Y = 30” tests compared with that of type 310 stai nl ess steel tested in the same conditi ons in nitrogen where a continuous scale did not form. 1. I ntroduction The surface degradation of metal s that occurs in aggressive environ- ments containing both corrosive and erosive media has been an important design consideration in the constr uction of equipment for several different industr ies. The loss of sound structural metal by erosion-corrosion can be experienced in such diverse equipment as gas turbines [l 1 and fluidized bed combustors [ 41. There are major differences in the operati ng environ- ments of key components in the two equipment examples referred to [5,6]. However, both the turbine blades in the gas turbine and the heat exchanger 0043-1648/86/$3.50 @ Elsevier Sequoia/Printed in The Netherlands
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/30/2019 Surface Degradation of Ductile Metals in Elevated
scale. This overcame the parabolic reaction rates which make the degradation
process much more time dependent in straight corrosion testing.
The data reported and analyzed in this paper are a part of a compre-
hensive program to investigate the elevated temperature behavior of a
number of alloys [ 8, lo]. The steels tested.in the total program were com-
mercial alloys commonly used in steam boiler and chemical process plant
components. Their designations are as follows: 1018; 2iCr-1Mo; 5Cr-+Mo;
SCr-1Mo; type 410 stainless steel; type 304 stainless steel; type 310 stainless
steel; 17-4PH. Their compositions are listed in ref. 10.
The two alloys reported on herein, type 310 stainless steel and 9Cr-
1Mo steel, behaved in a manner that was representative of all the alloys
investigated. The type 310 stainless steel was selected because it forms a
protective Cr,03 scale at the selected test conditions. The SCr-1Mo steelwas selected because its lower chromium content is marginal for forming a
protective scale at the test conditions. The erosion of the other alloys is
reported in ref. 10. They are comparatively simple alloys of iron having their
major variable element, chromium, in the range from 0 to 25 wt.%. This
range of chromium contents was selected because the oxide scales that form
on the metals in straight elevated temperature corrosion tests provide from
none to a fully protective Cr,Os scale.
The size of the specimens tested in the nozzle tester was 17.5 mm X
17.5 mm X 2 mm. The degradation rates of the test specimens were deter-
mined either by mass loss or thickness loss. To prevent oxidation of the testsurface prior to the tests in the elevated temperature erosion tester, undried
nitrogen was passed through the erosion tester until the specimen reached
the test temperature. After the test the specimen was quickly removed from
the furnace section of the tester and placed under a protective flow of
nitrogen until it had cooled to approximately 300 “C to prevent further
oxidation. Some spalling of the scale on the test surface occurred during
cooling. Optical and scanning electron microscopes were used to observe the
specimens’ surfaces and cross sections. Energy-dispersive X-ray analysis and
X-ray diffraction were used to determine the composition of the scales.
3. Results and discussion
3.1. Effect of t emperat ure on t he st rai ght erosi on of t ype 310 stai nl ess st eel
Figure 1 shows the erosion rate as a function of test temperature of
the highest chromium content most-corrosion-resistant steel tested, type
310 stainless steel [lo]. Each data point was obtained from a separately
tested specimen. The alloy was tested in a nitrogen carrier gas to prevent
corrosion from occurring. Four different test series were carried out to
determine the reproducibility of the data generated in the elevated tem-
perature erosion tester. At Q = 30” the erosion rate did not change as the
temperature was increased until 400 “C was reached. Above this temper-
ature the erosion rate increased with higher test temperatures at an increas-
ing rate.
7/30/2019 Surface Degradation of Ductile Metals in Elevated
Fig. 1. Erosion rate of type 310 stainless steel us. test temperature (20 pm Sic; u = 30m s-l) at (Y= 30” (A, 0, 0, 0) and (Y= 90” (A, m, 0, 4): A, A, run 1; 0, ., run 2; 0, 0, run 3;
0, +, run 4.
The shape of the curve for the CY 90” tests is somewhat different from
the cr = 30” curve. The erosion rate decreased from ambient temperature to
400 “C and then increased with temperature at an increasing slope. All the
alloys listed above showed this type of behavior with the decrease in erosion
rate at the lower elevated temperatures varying from essentially 0% to 60%less than the rate at room temperature.
The temperature at which the alloy steels started to undergo an increas-
ing erosion rate with test temperature correlated well with the temperature
at which their short-time tensile properties started to decrease at an increas-
ing rate. The decrease in erosion rate with test temperature at the lower
temperature showed the same trend as increases in the impact strength of
the alloys as the test temperature was increased above ambient temperature.
The reasons for these correlations are not known.
The effect of particle velocity on the erosion rate of type 310 stainless
steel tested at 800 “C is shown in Fig. 2. The velocity exponent of 1.23 isapproximately one-half of that reported for ductile metals at room temper-
ature [ll] in the range of velocities used in these tests. This indicates that
the relationship between the kinetic energy of the impacting particles andthe amount of material removed from the eroded surface that has been
observed in room temperature tests [ll] and modeled extensively is mod-
ified at elevated temperatures.
The erosion rate for the type 310 stainless steel in Figs. 1 and 2 for the
same test condition, particle velocity u = 30 m s-l at 800 “C, differs becauseof the impingement angle used. The erosion rate of 0.25 X lo-” g g-l in
Fig. 2 is greater than the value of 0.13 X 10m4g g-l in Fig. 1 because the data
in Fig. 2 were obtained at (x = 20”. This angle was nearer the peak rate im-pingement angle for type 310 stainless steel than was the (II= 30” angIe used
in the tests plotted in Fig. 1.
7/30/2019 Surface Degradation of Ductile Metals in Elevated
Fig. 3. Micrographs of an eroded type 310 stainless steel surface at various test temper-atures (nozzle tester; erosion; 240 pm Sic; u = 30 m s-l; t =: 30 min).
Figure 01 deg) T WI
3(a) 30 775
3& f 90 710
3(c) 30 397
3(d) 30 25
3(e) 90 25
(d)
7/30/2019 Surface Degradation of Ductile Metals in Elevated
Fig. 6. Effect of particle velocity on the scale morphology of SCr-1Mo steel in 5 h testsat (Y= 90” (nozzle tester; erosion-corrosion; 130 pm Al203; air; T= 850 “C; primaryzone): (a) u = 10 m s-l; (b) u = 30 m s-l; (c) u = 45 m s-l; (d) IJ = 70 m s-l.
consolidate the scale and the segmented domain type of morphology re-
sulted. At a particle velocity of 30 m s-i a transition in the morphology can
be seen. The scale still has segments, but they appear to be more densified
with smoother surfaces. In the 45 m s-l test the distinctly separated domainshave essentially disappeared and in the 70 m s-l test there is no evidence that
segmented domains remain. The differences in the morphology discussed
can be clearly seen in the original glossy micrographs, but in the printed
photographs these important differences are much harder to see.
7/30/2019 Surface Degradation of Ductile Metals in Elevated
The effect of the differences in the scale morphology in erosion-corro-
sion tests as a function of velocity and temperature on metal loss rates is
seen in Fig. 7 and Fig. 8 respectively. Figure 7 plots the metal thickness lossas a function of velocity at two impingement angles: Q!= 90’ and (x = 30”.
The (Y= 90” curve is a classic S-shaped transition curve, indicating that a
marked difference occurred in the erosion-corrosion mechanism in the
velocity region around v = 30 m s-l. Below the transition velocity, the
scale was eroded by a comparatively slow mechanism of chipping of small
pieces of scale [ 171. Above the transition velocity, the scale was removed
by a much faster mechanism, the periodic spalling of relatively large pieces
of scale [ 81. The scale loss rates on the corrosion-dominated surface trans-
lated into the underlying metal loss rates that were measured by an optical
micrometer on a specimen cross section, The curve for the CY 30” tests
will be discussed later.
It is thought that the change in thickness loss rate of the SCr-1Mo steel
at the higher particle velocities in the Q!= 90” tests is due to the change in
the manner in which the scale is removed rather than because of a change
from primarily corrosion to a synergistic combined erosion-corrosion
mechanism. Corrosion was observed as the dominant mechanism on all
surfaces of the specimens in all the tests at all velocities. The sharp increase
IO 20 so 40 so 80 70 80
Vek@ty of Pdcle m/s
I
I
760 800 860 900
Temperature 'C
Fig. 7. Effect of particle velocity on the metal thickness loss of SCr-1Mo steel in TE850 “C tests at a! = 30” (A) and a! = 90” (0) (130 pm Al,O,; air; t = 5 h).
Fig. 8. Effect of test temperature on the metal thickness loss of $Cr-1Mo steel in u = 70ms-1testsat&=900(130~Alz0~;air;t=5h).
7/30/2019 Surface Degradation of Ductile Metals in Elevated
in metal loss could not be the result of a sudden increase in erosivity of the
particles between 30 and 40 m s-i because the erosivity is a function of the
kinetic energy which increases uniformly as a function of the velocity (see
Fig. 2). The initiation of loss of larger pieces of scale by spalling at u = 30 ms-l was observed and this is thought to account for the increased metal loss.
The difference between the scale loss mechanism at the lower andhigher velocities is postulated to be due to the difference between the scale
morphologies. As seen in Fig. 6, at about u = 30 m s-l the force of the im-
pacting particles begins to densify and consolidate the segmented domainsof scale which occurred at the lower velocities. In its more continuous form
at the higher velocities, the scale can develop sufficiently high internal stresslevels to cause periodic spalling. At the lower velocities the scale’s separate
domains prevent these high stress levels from occurring and scale removalcan only occur by the chipping mechanism.The behavior is thought to be similar to that of plasma-sprayed thermal
barrier ceramic coatings on gas turbine components. They are purposelymicrocracked during application to reduce thermal fatigue failures by coat-
ing spallation [ 18,191. The large number of subcritical microcracks reducethe elastic modulus and, therefore, minimize the stresses that can develop
in the coating layer for a given strain level. This results in a strain accom-modation that reduces the spalling tendency of the coatings.
3.4. Protective scaleOn the right-hand ordinate in Fig. 7 the approximate mass loss of the
specimen was plotted. It was calculated from the eroded area, the metal thick-ness loss and the metal density. It is approximate because the contour of the
eroded area varies somewhat. Comparing the erosion rates for the type 310
stainless steel in Fig. 1 at 800 - 850 “C with the rates plotted in Fig. 7 for the
SCr-1Mo steel shows that the rates based on mass loss were the same at(Y= 90” and 3.0 X lop6 g g-l. At CY 30” the type 310 stainless steel had an
erosion rate that was five times that of the SCr-1Mo steel.There are several factors regarding the erosion behavior of the two
steels that indicate that the corrosion scale that formed on the SCr-1Mosteel in the 850 “C test could have provided some protection to the basemetal in the small-angle (a = 30”) tests.
(1) The type 310 stainless steel was tested in a nitrogen gas atmosphereand did not form a continuous corrosion scale.
(2) The general erosion behavior of brittle material such as the com-paratively thick continuous scale which formed on the SCr-1Mo steel is toundergo their highest erosion rate at a! = 90” and, by comparison, to be muchmore erosion resistant at o = 30”.
(3) In other investigations [ 201 it has been observed that austeniticstainless steels are more erosion resistant than ferritic steels.
(4) The tensile strength of the type 310 stainless steel at 850 “C is17.5 kgf mmP2 (25000 lbf in-‘) while that of the SCr-1Mo steel is 7 kgfmmm2 (10 000 Ibf inM2).
7/30/2019 Surface Degradation of Ductile Metals in Elevated
mat eri al s, Berkeley, CA, Januar y 1979, National Association of Corrosion Engineers,Houston, TX, 1979, pp. 139 - 173.
D. R. Spriggs and R. P. Brobst, The effect of PFBC particulate on the high velocity
erosion-corrosion of gas turbine materials,Proc. NACE Conf on Corrosion -Erosion
of Coal Conversi on System M at eri al s, Berkeley, CA, Januar y 1982, National Associa-tion of Corrosion Engineers, Houston, TX, 1982, pp. 799 - 831.
J. B. Gilmour, D. C. Briggs, A. Sui and H. Hindam, Preliminary results of the CanadianAFBC materials test, NACE Corrosion’ 85, Boston, M A, M arch 1985, NationalAssociation of Corrosion Engineers, Houston, TX, 1985, Paper 341.
M. F. Hussein and W. Tabakoff, Dynamic behavior of solid particles suspended bypolluted flow in a turbine state, J. A i rcr. 10 (7) (1973) 434 - 440.S. Jansson, Erosion and erosion-corrosion in fluidized bed combustor systems, hoc.NACE Conf. on Corrosi on-Erosion-Wear i n Emergi ng Fossi l Energy Syst ems, Berk el ey,
CA, Januar y 1982, National Association of Corrosion Engineers, Houston, TX, 1982,pp. 548 - 560.
H. Hindam and D. P. Whittle, Corrosion behavior of CraOs former alloys in Hz-H+H20 atmospheres, NA CE Corrosion’ 81, Toront o, Apr i l 1981, National Associa-tion of Corrosion Engineers, Houston, TX, 1981, Paper 93.A. V. Levy and Y.-F. Man, Elevated temperature erosion-corrosion of 9CrlMo steel,
Proc. AI M E Conf. on Hi gh Temperat ure Corrosion in Energy Systems, D et roi t , MI ,
Sept ember 1984, Metallurgical Society of AIME, New York, 1984.D. M. Kliest, One dimensional, two phase particulate flow, Rep. LBL -6967, 1977
(Lawrence Berkeley Laboratory, University of California, Berkeley, CA); M.S. Thesis,
University of California, Berkeley, CA, 1977.A. V. Levy, J. Yan and J. Patterson, Elevated temperature erosion of steels, Proc. Int.
Conf. on Wear of M at eri al s, Vancouver, Apr i l 1985, American Society of MechanicalEngineers, New York, 1985, in Wear, 108 (1) (1986) 43 - 60.
I. Finnie and D. H. McFadden, On the velocity dependence of the erosion of ductilemetals by solid particles at low angles of incidence, Wear, 48 (1) (1978) 181.
A. V. Levy, The erosion of metal alloys and their scales, Proc. NACE Conf. on Cor-
rosion- Erosi on-W ear of M at eri al s i n Emergi ng Fossi l Energy Systems, Berkel ey, CA,
January 1982, National Association of Corrosion Engineers, Houston, TX, 1982, p.
298.A. V. Levy, M. Aghazadeh and G. Hickey, The effect of test variables on the platelet
mechanism of erosion, Wear , 108 (1) (1986) 23 - 4 1.A. V. Levy, E. Slamovich and N. Jee, Elevated temperature combined erosion-corrosionof steels, Wear, 110 (2) (1986) 117 - 149.A. V. Levy and Y.-F. Man, The effect of temperature on the erosion-corrosion of
SCk-1Mo steel, NA CE Corrosion’ 85, Boston, M A, M arch 1985, Paper 337, in Wear,111 (1986) 161 - 172.A. V. Levy and Y.-F. Man, Elevated temperature erosion-corrosion of SCr-1Mo steel,
Wear, 111 (1986) 135 - 159.G. Zambelli and A. V. Levy, Particulate erosion of NiO scales, Wear, 68 (3) (1981)
305 - 331.T. E. Strangman, Thermal barrier coatings for turbine airfoils, AVS Int. Conf. on
Met all urgical Coati ngs, San Di ego, CA, Apri l 9 - 13, 1984, in Thin Solid Films, 127
(1985) 93 - 105.
I . E. Sumner and D. Ruckle, Development of improved durability plasma sprayedceramic coatings for gas turbine engines, AZ AA Paper 80-l 193, 1980 (AmericanInstitute for Astronautics and Aeronautics).
T. Foley and A. V. Levy, The erosion of heat-treated steels, Wear, 91 (1) (1983) 45 - 64.